New energy power generation grid-connected protocol conversion control method

Abstract

The invention relates to the field of new energy power generation control, in particular to a new energy power generation grid-connected protocol conversion control method. According to the method, the voltage, frequency and communication state of a grid-connected point are monitored in real time, and a support vector machine model is combined to judge whether a system is in a grid loss state; after network loss, the controller is switched to an isolated network operation protocol, a collaborative inertia regulation and control strategy is executed, output of each power generation unit and the energy storage system is dynamically distributed, and frequency and voltage stabilization is achieved; further adjusting the output of each new energy unit based on a linear quadratic Gaussian control algorithm to realize active power balance under the isolated network; after the power supply of the power grid is recovered, the frequency, voltage and phase synchronization of the power generation system and the power grid is realized by adopting self-adaptive fuzzy control through a set synchronization threshold judgment condition; and finally, combining a soft switching-on technology to realize safe grid connection and switching to a grid connection control protocol. The method has the advantages of quick response, high control precision, stable grid connection process and the like, and is suitable for various new energy microgrid scenes.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy power generation control, and in particular to a renewable energy power generation grid-connected protocol conversion control method, which is applicable to grid-connected dispatching control scenarios of various renewable energy systems such as wind power, photovoltaic, and electrochemical energy storage. Background Art

[0002] As the proportion of renewable energy access continues to increase, new energy systems face numerous challenges during grid-connected operation. When a fault or line trip occurs in the main grid, a new energy power station may become disconnected from the main grid, creating an isolated network. In this state, the power source must independently maintain local voltage and frequency stability. This isolated operation is not subject to centralized grid control and is prone to frequency and voltage fluctuations, which can impact equipment safety and power supply quality.

[0003] After the power grid is restored, new energy systems must be reconnected to the main grid, completing the transition from isolated to grid-connected operation. This process involves multiple control steps, including protocol switching, synchronization control, and circuit breaker closing judgment. Existing technologies focus primarily on normal grid-connection control strategies, but pay insufficient attention to the dynamic protocol conversion mechanism during the entire process of grid loss and reconnection. This lack of systematic and reliable control methods can easily lead to grid connection failures, current surges, and even equipment damage.

[0004] Therefore, a new energy power generation grid-connected protocol conversion control method is proposed. Summary of the Invention

[0005] The present invention provides a new energy power generation grid-connected protocol conversion control method, which enables the new energy system to enter isolated grid operation and stable control after the grid is lost, determines whether it has the grid connection conditions after the main grid is restored, and completes the smooth transition from the isolated grid control protocol to the grid-connected control protocol, thereby improving the safety, reliability and intelligence level of the system operation.

[0006] To achieve the above object, the present invention provides the following technical solutions: A new energy power generation grid-connected protocol conversion control method, comprising: Real-time monitoring of the operating status of the power generation system and determination of grid-off status using a support vector machine. When the power generation system is determined to be in a grid-off state, the current control strategy is switched to an isolated grid operation control protocol. Execute the isolated grid operation control protocol, call the coordinated inertia control strategy, and instantaneously adjust the output voltage and frequency of the power generation system according to the frequency change rate; Based on local load change information, the remaining capacity of the energy storage system, and system disturbance parameters, the linear quadratic Gaussian control algorithm dynamically adjusts the output power of each renewable energy generation unit to achieve active power balance in the isolated grid system. After detecting that the power grid has resumed power supply, determine whether the frequency difference, voltage difference and phase difference between the current isolated grid system and the main grid are within the set synchronization threshold range; When the synchronization conditions are met, the power generation system output frequency, voltage amplitude and phase are controlled to gradually match the grid parameters, and the electrical synchronization process before grid connection is completed through adaptive fuzzy control; After the electrical synchronization is completed, the grid-connected switch closing action is triggered, and the control protocol is switched from the isolated grid control protocol to the grid-connected operation protocol. The grid dispatching command is continued to be executed to complete the grid connection of the power generation system with the main grid.

[0007] Furthermore, the steps for determining the network loss state include: Collect the operating characteristic data of the power generation system grid connection point in real time and build the operating characteristic data into input vectors; A training set is constructed based on historical operating condition samples, and the optimal discrimination boundary is formed by training the support vector machine model to distinguish between grid connection, grid loss and abnormal transition states; The input vector is classified by the support vector machine model to obtain the classification result of the current state of the power generation system.

[0008] Furthermore, the execution steps of the collaborative inertia control strategy include: Real-time monitoring of the frequency change rate of the power generation system. Based on the frequency change rate, a gradient allocation function is called to dynamically allocate the power output ratio between the energy storage system and multiple power generation units. A virtual inertia weight matrix is ​​set as the control reference input parameter of each power generation unit. The local controller of each power generation unit dynamically adjusts its output voltage and frequency according to the weight matrix to achieve multi-source coordinated control.

[0009] Furthermore, the execution steps of the quadratic Gaussian control algorithm include: Constructing a linear state space model of the isolated grid system, the model including state equations of power generation unit frequency, voltage, and power output; Establish a performance index function and use the Kalman filter to optimally estimate the system state to obtain the state estimation value; Solve the Riccati equation, obtain the optimal feedback gain matrix, and construct the control law; The power output of each renewable energy power generation unit is adjusted in real time according to the state estimation value to achieve active power balance and frequency and voltage stability control of the isolated grid system.

[0010] Furthermore, the process of determining the synchronization threshold range includes: Monitor the voltage amplitude, frequency and phase information of the main grid in real time and compare them with the electrical parameters of the current isolated grid system; Calculate the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, and set a set of predefined synchronization threshold ranges. When the frequency difference, voltage difference and phase difference mentioned above all continue to meet the set threshold range for the set duration, it is determined that the grid connection conditions are met; otherwise, the isolated grid control state is maintained and the timing is restarted to wait for the next determination cycle.

[0011] Furthermore, the execution steps of the adaptive fuzzy control include: Collect the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, as well as the corresponding difference change rate, as control input data; Convert the control input data into fuzzy linguistic variables through membership functions; Reasoning based on the preset fuzzy control rule base to determine the control strategy for adjusting the output frequency, voltage and phase; The specific control output is generated through the defuzzification process to drive the inverter to complete parameter adjustment.

[0012] Furthermore, when executing the grid-connected switch closing operation, the zero-crossing point of the main grid voltage is detected and predicted, and a closing command is issued in advance so that the circuit breaker is closed at the zero-crossing point.

[0013] The beneficial effects of the present invention are: Through a collaborative inertia control strategy, during the initial transition from grid-connected to isolated power generation, multiple power generation units and the energy storage system can achieve a coordinated response, effectively suppressing frequency fluctuations and voltage fluctuations caused by grid loss. This strategy utilizes virtual inertia and a gradient allocation mechanism to achieve transient optimal output configuration for each power source, improving the system's robustness to sudden disturbances and dynamic stability.

[0014] Through the linear quadratic Gaussian control algorithm, during the stable operation phase of the isolated grid system, based on the optimal estimation of the system state and the minimum cost function control mechanism, the power output of each renewable energy power generation unit is dynamically adjusted, fully considering multi-dimensional information such as load changes, energy storage capacity and system disturbances, to achieve active power balance and voltage and frequency stability control within the isolated grid system, thereby effectively improving the overall energy management efficiency and system economy during the isolated grid operation period.

[0015] Through adaptive fuzzy control, during the electrical synchronization stage of the isolated grid system's reconnection, the error information and its changing trend are integrated to achieve fine adjustment of the output frequency, voltage and phase. The control rules can be adaptively adjusted according to the system's operating status to ensure the continuity and smoothness of the synchronization process, reduce grid connection shocks and transition disturbances, and improve the success rate and safety of the system's reclosing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a new energy power generation grid-connected protocol conversion control method provided by the present invention. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0018] Example 1 A new energy power generation grid-connected protocol conversion control method, such as Figure 1 Shown, including: S100: Real-time monitoring of the operating status of the power generation system and determination of the grid-off status using a support vector machine. When it is determined that the power generation system is in the grid-off state, the current control strategy is switched to the isolated grid operation control protocol; Furthermore, the steps for determining the network loss state include: Collect the operating characteristic data of the power generation system grid connection point in real time and build the operating characteristic data into input vectors; A training set is constructed based on historical operating condition samples, and the optimal discrimination boundary is formed by training the support vector machine model to distinguish between grid connection, grid loss and abnormal transition states; The input vector is classified by the support vector machine model to obtain the classification result of the current state of the power generation system.

[0019] Specifically, key operating characteristic parameters of the power generation system grid connection point are collected in real time, including at least but not limited to: three-phase voltage effective value, three-phase frequency, voltage and frequency change rate, active power and reactive power, and communication connection status signs (such as heartbeat signal, data packet loss rate). After standardized preprocessing, the above data are constructed into a multi-dimensional input vector as the input of the model.

[0020] Based on historical operating data, a known labeled sample set is constructed. The sample labels include State A: normal grid connection; State B: grid loss; State C: abnormal transition (frequency jitter, edge state, etc.). Feature extraction and data cleaning are performed on these samples, and the support vector machine is trained using a kernel function (such as the radial basis kernel function (RBF)) to obtain the optimal hyperplane or discriminant boundary that can distinguish the three types of states.

[0021] During system operation, the controller periodically collects current operational characteristic data, constructs it into an input vector, and feeds it into a trained support vector machine model. The SVM model outputs a classification result for this input vector, representing the current system state. If it is identified as "State A," grid-connected operation is maintained; if it is identified as "State B," the controller is triggered to execute the grid-connected protocol and switch to isolated operation; if it is identified as "State C," it can be further determined to be in a transitional state, entering buffering or continuing monitoring.

[0022] The use of support vector machines for network loss judgment can effectively identify system disconnections caused by faults, communication loss, sudden interference, etc., while taking into account the two indicators of "low false detection rate" and "fast response speed", providing an intelligent foundation for subsequent protocol switching and improving the stability and autonomous control capabilities of new energy systems.

[0023] S200: Execute the isolated grid operation control protocol, call the coordinated inertia control strategy, and instantaneously adjust the output voltage and frequency of the power generation system according to the frequency change rate; Furthermore, the execution steps of the collaborative inertia control strategy include: Real-time monitoring of the frequency change rate of the power generation system. Based on the frequency change rate, a gradient allocation function is called to dynamically allocate the power output ratio between the energy storage system and multiple power generation units. A virtual inertia weight matrix is ​​set as the control reference input parameter of each power generation unit. The local controller of each power generation unit dynamically adjusts its output voltage and frequency according to the weight matrix to achieve multi-source coordinated control.

[0024] Specifically, the system controller monitors the rate of change of the local frequency of the power generation system in real time, and uses a high-resolution frequency sampling module to perform millisecond-level detection of the system frequency to ensure that the frequency change trend caused by load disturbances or power generation fluctuations can be captured in a timely manner.

[0025] The controller introduces a gradient distribution function Assign an output adjustment ratio to each response unit. The function form is as follows: ; in, Indicates the The capacity factor of each power generation unit / energy storage device, Indicates the response sensitivity parameter (between 0.5 and 2), The output is The share of inertial response power that each unit should bear, Indicates the frequency change rate. After normalization, the power allocation weight coefficient is formed , each unit is based on To adjust its output power, It is the active power regulation quantity that the system needs to respond to.

[0026] According to the above weights , the controller constructs a virtual inertia weight matrix , which has the form: ; in, Indicates the The equivalent virtual inertia coefficient of each element. Matrix It is sent to each local controller as a reference input, and the controller updates the local voltage and frequency control targets according to the following formula: ; ; in, Indicates the updated frequency, represents the reference frequency, Indicates the updated voltage, represents the reference frequency, and Indicates that according to The local controller controls the inverter output voltage and frequency of the corresponding power generation equipment according to the above target values.

[0027] This collaborative inertia control strategy, combined with a gradient allocation function and a virtual inertia weight matrix, not only improves the power generation system's rapid response and frequency support capabilities in the event of a grid outage, but also enables coordinated and stable control among multiple power generation units. Compared to traditional droop or single-point control schemes, this strategy offers greater dynamic adaptability and coordination, making it particularly suitable for isolated grid operation with high penetration of renewable energy.

[0028] S300: Based on local load change information, the remaining capacity of the energy storage system, and system disturbance parameters, it dynamically adjusts the output power of each renewable energy generation unit through a linear quadratic Gaussian control algorithm to achieve active power balance in the isolated grid system. Furthermore, the execution steps of the quadratic Gaussian control algorithm include: Constructing a linear state space model of the isolated grid system, the model including state equations of power generation unit frequency, voltage, and power output; Establish a performance index function and use the Kalman filter to optimally estimate the system state to obtain the state estimation value; Solve the Riccati equation, obtain the optimal feedback gain matrix, and construct the control law; The power output of each renewable energy power generation unit is adjusted in real time according to the state estimation value to achieve active power balance and frequency and voltage stability control of the isolated grid system.

[0029] Specifically, a simplified linear dynamic model of the isolated grid system is established. The model uses key variables such as system frequency, voltage, and active power output of each renewable energy generation unit as state variables to construct the state equation and output equation of the state space. The state equation and output equation are: ; ; in, Represents the system state vector, such as frequency deviation, voltage amplitude, and output of each power generation unit, Represents the control input vector, corresponding to the power regulation of each power generation unit, Represents measurable output variables, such as frequency, voltage, 、 and represents the system dynamic matrix, and represents process noise and measurement noise, which satisfy zero-mean Gaussian distribution. It is worth noting that matrix 、 and It can be obtained by methods such as recursive least squares, support vector regression or LSTM. Solving the system dynamic matrix is ​​a conventional technical means and will not be described here.

[0030] Since some system states cannot be measured directly, a Kalman filter is used to estimate the state in real time. The filter predicts the state and corrects the error to achieve the optimal estimate of the state variables. The process of predicting the state and correcting the error can be expressed as: ; ; in, Represents time steps The estimated state value at represents the Kalman gain, which is recursively updated according to the estimated error covariance.

[0031] In order to optimize the performance of the control system, the following quadratic objective function is constructed: ; in, is the state weight matrix, which measures the degree of penalty for the system deviating from the stable state. The input weight matrix is ​​used to control the energy cost of regulation.

[0032] By solving the discrete-time algebraic Riccati equation, the feedback gain matrix is ​​obtained , and establish the control law This control law associates the estimated state with the control input to achieve optimal closed-loop regulation of the system. Based on this control law, the controller outputs real-time power commands for each renewable energy generation unit. It dynamically allocates power based on the remaining capacity of the energy storage system and local load demand, achieving stable voltage and frequency control and active power balance in the isolated grid under disturbances and load changes.

[0033] The use of the linear quadratic Gaussian control algorithm can effectively suppress frequency and voltage fluctuations caused by load changes and external disturbances during isolated grid operation; at the same time, it can achieve the optimal distribution of renewable energy output among multiple sources. It has the advantages of fast response, high control accuracy, and strong robustness, significantly improving the stability and safety of isolated grid operation.

[0034] S400: After detecting that the power grid has resumed power supply, determining whether the frequency difference, voltage difference, and phase difference between the current isolated grid system and the main grid are within a set synchronization threshold range; Furthermore, the process of determining the synchronization threshold range includes: Monitor the voltage amplitude, frequency and phase information of the main grid in real time and compare them with the electrical parameters of the current isolated grid system; Calculate the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, and set a set of predefined synchronization threshold ranges. When the frequency difference, voltage difference and phase difference mentioned above all continue to meet the set threshold range for the set duration, it is determined that the grid connection conditions are met; otherwise, the isolated grid control state is maintained and the timing is restarted to wait for the next determination cycle.

[0035] Specifically, the control system collects the main grid's electrical operating parameters in real time, including voltage amplitude, frequency, and voltage phase information. It also collects the corresponding parameters of the current isolated grid system, namely the voltage amplitude, frequency, and phase of the isolated grid itself.

[0036] The control system compares three key electrical parameters between the isolated grid system and the main grid, calculating the frequency, voltage, and phase differences. For example, the frequency difference is the absolute difference between the main grid frequency and the isolated grid system frequency. The voltage difference can be evaluated using a relative error method, and the phase difference is the difference in phase angle between the two voltage waveforms.

[0037] To ensure safe grid connection, the system pre-sets a set of synchronization thresholds, including but not limited to the following parameter ranges: frequency difference within ±0.1Hz, voltage amplitude difference within ±10%, and phase difference within ±10 degrees. These thresholds can be adjusted based on the grid voltage level, grid characteristics, and inverter tolerance.

[0038] To avoid misjudgments due to transient disturbances or short-term anomalies, the control system sets a judgment duration threshold. The system is considered eligible for grid connection only when the differences between the three parameters consistently meet the preset synchronization threshold for a period of time (e.g., 2 seconds). Otherwise, the system maintains isolated grid control mode and restarts the timing for the next judgment round.

[0039] When the above conditions are continuously met, the system outputs a synchronization completion signal and enters the grid connection preparation state. Subsequently, operations such as electrical synchronization adjustment and circuit breaker closing can be performed.

[0040] Through the above-mentioned determination process, the risk of grid-connected electrical shock can be effectively reduced, ensuring that the new energy power generation system has a sufficient synchronization basis when it is reconnected to the grid.

[0041] S500: When the synchronization conditions are met, the power generation system output frequency, voltage amplitude and phase are controlled to gradually match the grid parameters, and the electrical synchronization process before grid connection is completed through adaptive fuzzy control; Furthermore, the execution steps of the adaptive fuzzy control include: Collect the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, as well as the corresponding difference change rate, as control input data; Convert the control input data into fuzzy linguistic variables through membership functions; Reasoning based on the preset fuzzy control rule base to determine the control strategy for adjusting the output frequency, voltage and phase; The specific control output is generated through the defuzzification process to drive the inverter to complete parameter adjustment.

[0042] Specifically, the controller collects the differences in three key electrical parameters between the isolated grid system and the main grid in real time, including: frequency error, voltage amplitude error, phase error, and the rate of change of the above differences, to form a control input vector.

[0043] The collected error values ​​and their rates of change are input into a pre-set membership function module and converted into corresponding fuzzy linguistic variables. For example, frequency error can be categorized into {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}; the error rate of change is also categorized into corresponding levels. Membership functions can be triangular, trapezoidal, or Gaussian, with the specific categorization range determined by the grid level and control accuracy requirements.

[0044] Design a fuzzy control rule base for frequency, voltage, and phase. The rules are expressed as an "if-then" structure. For example, if the frequency error is small and positive, and the frequency error rate of change is moderately negative, then the output adjustment is "decrease slightly"; if the voltage amplitude error is large and negative, and the voltage amplitude error rate of change is large and positive, then the output is "increase rapidly." This rule base can be built based on expert experience or simulation models and include a self-learning mechanism to dynamically update rule weights.

[0045] The controller performs logical reasoning on the current fuzzy input based on a fuzzy reasoning mechanism (such as the Mamdani or Sugeno model) to obtain the fuzzy control output result; then, the fuzzy output is converted into a specific numerical control quantity through a defuzzification process (such as the center of gravity method or the maximum membership method).

[0046] The controller sends the defuzzified numerical control signal to the inverter or converter control module, driving the power generation unit to adjust its output frequency, voltage and phase so that it gradually approaches the grid parameters and realizes the electrical synchronization process before grid connection.

[0047] By introducing the adaptive fuzzy control method, flexible electrical synchronization regulation between the new energy power generation system and the main power grid is achieved under complex disturbances and nonlinear working conditions, effectively improving the frequency, voltage and phase synchronization accuracy, reducing the impact risk during the grid connection process, and having good dynamic response capability and robustness, ensuring that the power generation system is smoothly and safely connected to the main power grid.

[0048] S600: After the electrical synchronization is completed, the grid-connected switch closing action is triggered, and the control protocol is switched from the isolated grid control protocol to the grid-connected operation protocol. The grid dispatching command is continued to be executed to complete the grid connection of the power generation system with the main grid.

[0049] Furthermore, when executing the grid-connected switch closing operation, the zero-crossing point of the main grid voltage is detected and predicted, and a closing command is issued in advance so that the circuit breaker is closed at the zero-crossing point.

[0050] Specifically, after the power generation system synchronizes its electrical parameters (frequency, voltage, and phase) with the main grid, the control system enters a grid-connection preparation state. To reduce the inrush current at the moment of closing, the system samples the main grid voltage waveform in real time and extracts its zero-crossing information. A zero-crossing detection algorithm is used to predict the arrival time of the next voltage zero-crossing point. Based on the mechanical delay characteristics of the circuit breaker closing, the controller issues a closing command in advance to ensure that the circuit breaker physically closes at the moment the main grid voltage crosses zero.

[0051] After the closing is completed, the controller immediately switches the control protocol, smoothly transitioning from the isolated grid operation protocol to the grid-connected operation protocol, reactivating grid-connected control strategies such as PQ control, constant power control or AGC tracking, responding to grid dispatch commands, and ensuring that the new energy power generation system and the main grid are connected to the grid safely and reliably on the basis of consistent electrical parameters.

[0052] Example 2 This embodiment is applicable to an independent microgrid system that includes wind power generation, photovoltaic power generation, and energy storage systems. The system has the dual capabilities of autonomous operation and grid-connected operation, and is used for power supply guarantee in industrial parks or remote areas. In a certain microgrid system scenario, the implementation of a new energy power generation grid-connected protocol conversion control method proposed in the present invention includes: Microgrids use front-end intelligent monitoring devices (such as PMUs and smart sensors) to collect real-time data on voltage, frequency, and communication status at the grid connection point. This data is constructed into feature vectors and fed into a pre-trained support vector machine classification model. The model classifies the input vectors and identifies whether the system is currently connected to the grid, disconnected from the grid, or experiencing an abnormal transition. If the system is determined to be disconnected from the grid, the control center immediately issues a control command to switch to the isolated grid operation protocol.

[0053] The microgrid controller activates a coordinated inertia control strategy. This strategy uses the system frequency rate as the core control variable and invokes a preset gradient allocation function to dynamically adjust power distribution among wind turbines, photovoltaic inverters, and energy storage units. Each power generation unit's local controller adjusts parameters based on a predefined "virtual inertia weight matrix" to achieve rapid frequency and voltage stabilization.

[0054] During isolated grid operation, the main controller constructs a state-space model of the microgrid system based on collected data on local load changes, energy storage SOC, and disturbances. It employs a linear quadratic Gaussian control algorithm, employs a Kalman filter to estimate system states, and implements a power control law in conjunction with an optimized feedback gain matrix to dynamically allocate the output power of each renewable energy unit. This process ensures active power balance within the microgrid and maintains voltage and frequency within safe ranges.

[0055] When the system detects that the external grid has restored power, the controller activates the main grid parameter acquisition module, acquiring real-time frequency, voltage, and phase data. This data is then compared with the microgrid's current operating parameters. If the frequency, voltage, and phase differences are all within the set thresholds and remain within these thresholds for a set period (e.g., 5 seconds), the conditions for grid connection are determined to be met.

[0056] The system uses an adaptive fuzzy control strategy to process the differences between the three parameters and their rates of change. After converting them into fuzzy linguistic variables, the system then generates an electrical regulation strategy based on a fuzzy rule base. After defuzzification, these signals are generated to drive the outputs of the wind and photovoltaic inverters, gradually aligning their voltage, frequency, and phase with the grid for synchronization.

[0057] The controller predicts the zero-crossing point of the main grid voltage in real time and issues a closing command in advance. The circuit breaker closes at this point, effectively suppressing the inrush current. After closing, the system immediately switches to the grid-connected control protocol, enabling constant power control or PQ control, responding to main grid dispatch commands and completing the smooth connection of the microgrid.

[0058] This solution automatically controls the entire process of a new energy system in a microgrid environment, from grid loss to isolated operation to pre-grid synchronization and reconnection. It offers high robustness, low voltage surges, and strong grid security, making it suitable for hybrid microgrid systems with energy storage. It is particularly well-suited for scenarios where frequent grid fluctuations require rapid mode switching.

[0059] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A new energy power generation grid-connected protocol conversion control method, characterized in that: include: Real-time monitoring of the operating status of the power generation system and determination of grid-off status using a support vector machine. When the power generation system is determined to be in a grid-off state, the current control strategy is switched to an isolated grid operation control protocol. Execute the isolated grid operation control protocol, call the coordinated inertia control strategy, and instantaneously adjust the output voltage and frequency of the power generation system according to the frequency change rate; Based on local load change information, the remaining capacity of the energy storage system, and system disturbance parameters, the linear quadratic Gaussian control algorithm dynamically adjusts the output power of each renewable energy generation unit to achieve active power balance in the isolated grid system. After detecting that the power grid has resumed power supply, determine whether the frequency difference, voltage difference and phase difference between the current isolated grid system and the main grid are within the set synchronization threshold range; When the synchronization conditions are met, the power generation system output frequency, voltage amplitude and phase are controlled to gradually match the grid parameters, and the electrical synchronization process before grid connection is completed through adaptive fuzzy control; After the electrical synchronization is completed, the grid-connected switch closing action is triggered, and the control protocol is switched from the isolated grid control protocol to the grid-connected operation protocol. The grid dispatching command is continued to be executed to complete the grid connection of the power generation system with the main grid.

2. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: The steps for determining the network loss status include: Collect the operating characteristic data of the power generation system grid connection point in real time and build the operating characteristic data into input vectors; A training set is constructed based on historical operating condition samples, and the optimal discrimination boundary is formed by training the support vector machine model to distinguish between grid connection, grid loss and abnormal transition states; The input vector is classified by the support vector machine model to obtain the classification result of the current state of the power generation system.

3. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: The execution steps of the collaborative inertia control strategy include: Real-time monitoring of the frequency change rate of the power generation system. Based on the frequency change rate, a gradient allocation function is called to dynamically allocate the power output ratio between the energy storage system and multiple power generation units. A virtual inertia weight matrix is ​​set as the control reference input parameter of each power generation unit. The local controller of each power generation unit dynamically adjusts its output voltage and frequency according to the weight matrix to achieve multi-source coordinated control.

4. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: The execution steps of the quadratic Gaussian control algorithm include: Constructing a linear state space model of the isolated grid system, the model including state equations of power generation unit frequency, voltage, and power output; Establish a performance index function and use the Kalman filter to optimally estimate the system state to obtain the state estimation value; Solve the Riccati equation, obtain the optimal feedback gain matrix, and construct the control law; The power output of each renewable energy power generation unit is adjusted in real time according to the state estimation value to achieve active power balance and frequency and voltage stability control of the isolated grid system.

5. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: The synchronization threshold range determination process includes: Monitor the voltage amplitude, frequency and phase information of the main grid in real time and compare them with the electrical parameters of the current isolated grid system; Calculate the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, and set a set of predefined synchronization threshold ranges. When the frequency difference, voltage difference and phase difference mentioned above all continue to meet the set threshold range for the set duration, it is determined that the grid connection conditions are met; otherwise, the isolated grid control state is maintained and the timing is restarted to wait for the next determination cycle.

6. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: The execution steps of adaptive fuzzy control include: Collect the frequency difference, voltage difference and phase difference between the isolated grid system and the main grid, as well as the corresponding difference change rate, as control input data; Convert the control input data into fuzzy linguistic variables through membership functions; Reasoning based on the preset fuzzy control rule base to determine the control strategy for adjusting the output frequency, voltage and phase; The specific control output is generated through the defuzzification process to drive the inverter to complete parameter adjustment.

7. A new energy power generation grid-connected protocol conversion control method according to claim 1, characterized in that: When executing the grid-connected switch closing operation, the zero-crossing point of the main grid voltage is detected and predicted, and a closing command is issued in advance so that the circuit breaker is closed at the zero-crossing point.

Images